Automated Scoring of Alzheimer's Disease Atrophy Scale with Subtype Classification Using Deep Learning-Based T1-Weighted Magnetic Resonance Image Segmentation.

Background: Application of visual scoring scales for regional atrophy in Alzheimer's disease (AD) in clinical settings is limited by their high time cost and low intra/inter-rater agreement. Objective: To provide automated atrophy scoring using objective volume driven from deep-learning segmentation methods for AD subtype classification using magnetic resonance imaging (MRI). Methods: We enrolled 3,959 participants (1,732 cognitively normal [CN], 1594 with mild cognitive impairment [MCI], and 633 with AD). The occupancy indices for each regional volume were calculated by dividing each volume by the size of the lateral and inferior ventricular volumes. MR images from 355 participants (119 CN, 119 MCI, and 117 AD) from three different centers were used for validation. Two neuroradiologists performed visual assessments of the medial temporal, posterior, and global cortical atrophy scores in the frontal lobe using T1-weighted MR images. Images were also analyzed using the deep learning-based segmentation software, Neurophet AQUA. Cutoff values for the three scores were determined using the data distribution according to age. The scoring results were compared for consistency and reliability. Results: Four volumetric-driven scoring results showed a high correlation with the visual scoring results for AD, MCI, and CN. The overall agreement with human raters was weak-to-moderate for atrophy scoring in CN participants, and good-to-almost perfect in AD and MCI participants. AD subtyping by automated scores also showed usefulness as a research tool. Conclusions: Determining AD subtypes using automated atrophy scoring for late-MCI and AD could be useful in clinical settings or multicenter studies with large datasets.

A multimodal machine learning model for predicting dementia conversion in Alzheimer's disease.

Alzheimer's disease (AD) accounts for 60-70% of the population with dementia. Mild cognitive impairment (MCI) is a diagnostic entity defined as an intermediate stage between subjective cognitive decline and dementia, and about 10-15% of people annually convert to AD. We aimed to investigate the most robust model and modality combination by combining multi-modality image features based on demographic characteristics in six machine learning models. A total of 196 subjects were enrolled from four hospitals and the Alzheimer's Disease Neuroimaging Initiative dataset. During the four-year follow-up period, 47 (24%) patients progressed from MCI to AD. Volumes of the regions of interest, white matter hyperintensity, and regional Standardized Uptake Value Ratio (SUVR) were analyzed using T1, T2-weighted-Fluid-Attenuated Inversion Recovery (T2-FLAIR) MRIs, and amyloid PET (αPET), along with automatically provided hippocampal occupancy scores (HOC) and Fazekas scales. As a result of testing the robustness of the model, the GBM model was the most stable, and in modality combination, model performance was further improved in the absence of T2-FLAIR image features. Our study predicts the probability of AD conversion in MCI patients, which is expected to be useful information for clinician's early diagnosis and treatment plan design.

Magnetic resonance image-based brain age as a discriminator of dementia conversion in patients with amyloid-negative amnestic mild cognitive impairment.

Patients with amyloid-negative amnestic mild cognitive impairment (MCI) have a conversion rate of approximately 10% to dementia within 2 years. We aimed to investigate whether brain age is an important factor in predicting conversion to dementia in patients with amyloid-negative amnestic MCI. We conducted a retrospective cohort study of patients with amyloid-negative amnestic MCI. All participants underwent detailed neuropsychological evaluation, brain magnetic resonance imaging (MRI), and [18F]-florbetaben positron emission tomography. Brain age was determined by the volumetric assessment of 12 distinct brain regions using an automatic segmentation software. During the follow-up period, 38% of the patients converted from amnestic MCI to dementia. Further, 73% of patients had a brain age greater than their actual chronological age. When defining 'survival' as the non-conversion of MCI to dementia, these groups differed significantly in survival probability (p = 0.036). The low-educated female group with a brain age greater than their actual age had the lowest survival rate among all groups. Our findings suggest that the MRI-based brain age used in this study can contribute to predicting conversion to dementia in patients with amyloid-negative amnestic MCI.

Improvement of road safety management systems of local governments in Korea after evaluating related indicators.

This study analyzes what has changed in the road safety management systems (RSMS) of local governments after evaluating the related indicators continuously from 2015 to 2019. It has evaluated 24 indicators in four areas of RSMS such as institutional management, intervention, intermediate outcome, and the final outcome, according to Han and Lee (2020). It seems that continuous evaluation of indicators of RSMSs in local governments for five years has helped to improve road safety performance in local governments. The result shows that the average overall scores have increased from 77.92 to 83.89, a 7.7% increase. This improvement should have resulted from the competition between local governments, which do not want to be at the bottom in the score comparison. The most remarkable increase has been identified in the areas of institutional management and intervention. They have increased by 13.3% and 16.0%, respectively. In terms of types of local governments, it has been found that the increase in overall scores is higher in provinces which include most rural areas than in metropolitans. However, it has been found that the improvement of institutional management and interventions are not directly associated with the increase of the intermediate and final outcomes, at least during the five years. Further studies examining the relationship between the four areas of RSMS are needed.

A Fully Automated Visual Grading System for White Matter Hyperintensities of T2-Fluid Attenuated Inversion Recovery Magnetic Resonance Imaging.

BACKGROUND: The Fazekas scale is one of the most commonly used visual grading systems for white matter hyperintensity (WMH) for brain disorders like dementia from T2-fluid attenuated inversion recovery magnetic resonance (MR) images (T2-FLAIRs). However, the visual grading of the Fazekas scale suffers from low-intra and inter-rater reliability and high labor-intensive work. Therefore, we developed a fully automated visual grading system using quantifiable measurements. METHODS: Our approach involves four stages: (1) the deep learning-based segmentation of ventricles and WMH lesions, (2) the categorization into periventricular white matter hyperintensity (PWMH) and deep white matter hyperintensity (DWMH), (3) the WMH diameter measurement, and (4) automated scoring, following the quantifiable method modified for Fazekas grading. We compared the performances of our method and that of the modified Fazekas scale graded by three neuroradiologists for 404 subjects with T2-FLAIR utilized from a clinical site in Korea. RESULTS: The Krippendorff's alpha across our method and raters (A) versus those only between the radiologists (R) were comparable, showing substantial (0.694 vs. 0.732; 0.658 vs. 0.671) and moderate (0.579 vs. 0.586) level of agreements for the modified Fazekas, the DWMH, and the PWMH scales, respectively. Also, the average of areas under the receiver operating characteristic curve between the radiologists (0.80 ± 0.09) and the radiologists against our approach (0.80 ± 0.03) was comparable. CONCLUSIONS: Our fully automated visual grading system for WMH demonstrated comparable performance to the radiologists, which we believe has the potential to assist the radiologist in clinical findings with unbiased and consistent scoring.

Longitudinal follow-up of serum biomarkers in patients with neuromyelitis optica spectrum disorder.

BACKGROUND: Recently, several serum biomarkers have been proposed in Neuromyelitis Optica Spectrum Disorders (NMOSD) to monitor disease activity. OBJECTIVE: The objective of the study is to evaluate the longitudinal clinical value of serum biomarkers in patients with NMOSD. METHODS: We prospectively recruited consecutive NMOSD patients with anti-aquaporin-4 antibody and obtained serum samples at enrollment, after 6-12 months of follow-up (main period), and at attacks. Using single-molecule array assays, we evaluated longitudinal changes of serum neurofilament light chain (NfL), glial fibrillary acidic protein (GFAP), and GFAP/NfL levels. RESULTS: Overall, 64 patients (58 women) were enrolled (age: 51 years, disease duration: 6.7 years) and 133 samples were obtained. Among patients who did not develop new attacks during the main period (n = 62), serum levels of NfL, GFAP, and GFAP/NfL were significantly decreased over time in patients with attacks (<2 months) at enrollment (n = 14 (23%)), whereas serum NfL and GFAP levels gradually increased in the others (n = 48 (77%)). During the study, five (8%) patients developed new attacks; only serum GFAP levels increased consistently upon these events compared with baseline levels. To differentiate attacks from remissions, serum GFAP levels showed the largest area under the receiver operating characteristic curve (0.876, 95% confidence interval: 0.801-0.951). CONCLUSION: Among NfL, GFAP, and GFAP/NfL, serum GFAP might be the most appropriate for monitoring NMOSD longitudinally, which warrants future confirming studies.

Deep Learning-Based Method to Differentiate Neuromyelitis Optica Spectrum Disorder From Multiple Sclerosis.

Background: Differentiating neuromyelitis optica spectrum disorder (NMOSD) from multiple sclerosis (MS) is crucial in the field of diagnostics because, despite their similarities, the treatments for these two diseases are substantially different, and disease-modifying treatments for MS can worsen NMOSD. As brain magnetic resonance imaging (MRI) is an important tool to distinguish the two diseases, extensive research has been conducted to identify the defining characteristics of MRI images corresponding to these two diseases. However, the application of such research in clinical practice is still limited. In this study, we investigate the applicability of a deep learning-based algorithm for differentiating NMOSD from MS. Methods: In this study, we included 338 participants (213 patients with MS, 125 patients with NMOSD) who visited the Asan medical center between February 2009 and February 2020. A 3D convolutional neural network, which is a deep learning-based algorithm, was trained using fluid-attenuated inversion recovery images and clinical information of the participants. The performance of the final model in differentiating NMOSD from MS was evaluated and compared with that of two neurologists. Results: The deep learning-based model exhibited an area under the receiver operating characteristic curve of 0.82 (95% CI, 0.75-0.89). It differentiated NMOSD from MS with an accuracy of 71.1% (sensitivity = 87.8%, specificity = 61.6%), which is comparable to that exhibited by the neurologists. The intra-rater reliability of the two neurologists was moderate (κ = 0.47, 0.50), which was in contrast with the consistent classification of the deep learning-based model. Conclusion: The proposed model was verified to be capable of differentiating NMOSD from MS with accuracy comparable to that of neurologists, exhibiting the advantage of consistent classification. As a result, it can aid differential diagnosis between two important central nervous system inflammatory diseases in clinical practice.

Clinical implication of serum biomarkers and patient age in inflammatory demyelinating diseases.

OBJECTIVES: Serum synaptic proteins levels may change with age-related neurodegeneration, affecting their clinical implications as a disease biomarker. We aimed to investigate neuronal and astroglial markers in patients with multiple sclerosis (MS) and aquaporin-4 antibody-seropositive neuromyelitis optica spectrum disorders (NMOSD) to compare the clinical implications of these markers according to age. METHODS: Using single-molecule array assays, we measured neurofilament light (NfL) and glial fibrillary acidic protein (GFAP) in sera from consecutive patients with MS (n = 117) and NMOSD (n = 63). For each disease, we assessed correlations between these markers and disease severity (Expanded Disability Status Scale [EDSS]) scores according to three age groups (≤44, 45-54, and ≥55 years). RESULTS: Although serum GFAP levels were significantly higher in patients with NMOSD than those with MS, levels of both serum markers revealed significant positive correlations with EDSS scores in both diseases. In MS patients, the degrees of correlation between serum NfL (or GFAP) levels and EDSS scores were similar across all age groups. However, in NMOSD patients, positive GFAP-EDSS correlations were distinctively stronger in the youngest than in the oldest group. Conversely, there were no positive NfL-EDSS correlations in NMOSD in the youngest group, but there were significant in the oldest group. INTERPRETATION: The degrees to which serum NfL and GFAP levels reflect disease severity vary significantly with patient age in NMOSD, but not in MS. These findings suggest that the pathological processes and progression differ between the diseases; hence, serum biomarker levels may need to be interpreted differently according to patient age and disease type.

Beneficial effects of intravenous immunoglobulin as an add-on therapy to azathioprine for NMO-IgG-seropositive neuromyelitis optica spectrum disorders.

BACKGROUND: The efficacy of intravenous immunoglobulin (IVIG) in neuromyelitis optica spectrum disorders (NMOSD) has rarely been investigated, despite it being a conceivable therapeutic strategy based on its mode of action. We aimed to evaluate the efficacy of IVIG as an add-on therapy to azathioprine in the prevention of NMOSD relapse. METHODS: We retrospectively reviewed the medical records of NMO-IgG-positive NMOSD patients treated with IVIG infusions (0.4 g/kg/day) every one to three months as a part of combination therapy with azathioprine for more than six months. Treatment efficacy and safety were assessed based on the changes in the pre-IVIG and post-IVIG treatment annualized relapse rates (ARR), and on the proportion of relapse-free and progression-free patients and adverse events. RESULTS: This study was performed on 20 patients (19 women; median age 52 years). After add-on therapy with IVIG, 19 patients (95%) showed a significant reduction in the median ARR from 1.1 [interquartile range, 0.6‒1.4] to 0.3 [interquartile range, 0‒0.6] (p < 0.001). Seven patients (35%) were relapse-free during 43.5 months (median) of treatment. The median Expanded Disability Status Scale (EDSS) score remained stable at 4.0. In addition, 80% of the patients showed no disability progression, and 25% of the patients experienced an improvement in EDSS. The median NMO-IgG titer decreased from 1:480 (n = 19) to 1:120 (n = 13) after treatment (p = 0.006) and was found to be negative in three patients. Six patients (30%) stopped IVIG due to relapses after 27.5 months (median; interquartile range, 15.3‒43.3) of IVIG initiation. There were no severe side effects that led to discontinuation of IVIG. CONCLUSION: IVIG as add-on therapy may be associated with beneficial effects in preventing relapse and disability progression in NMO-IgG-positive NMOSD patients who have breakthrough disease activity despite immunosuppressive treatment with azathioprine. Further randomized controlled trials are necessary to validate the efficacy of IVIG in NMOSD.

Choroid plexus changes on magnetic resonance imaging in multiple sclerosis and neuromyelitis optica spectrum disorder.

OBJECTIVE: To investigate alterations in the choroid plexus in multiple sclerosis (MS) and neuromyelitis optica spectrum disorder (NMOSD) using brain magnetic resonance imaging (MRI). METHODS: We prospectively recruited consecutive patients with MS or NMOSD from July 2018 to February 2019. The inclusion criterion was brain MRI within three months from onset of acute neurological symptoms. The thickness and enhancement ratio of the choroid plexus on gadolinium-enhanced T1-weighted images of patients with MS (n = 51), patients with NMOSD (n = 32), and healthy controls (HCs, n = 28) were compared. RESULTS: MRI in patients with MS or NMOSD showed a comparably thick but more enhanced choroid plexus compared with that of HCs. In the axial view, enhancement ratios of the lateral ventricle of MS and NMOSD patients and HCs were 1.64 ± 0.34, 1.65 ± 0.25, and 1.39 ± 0.17, respectively (P > .999 for MS vs. NMOSD; P = .001 for MS vs. HCs; P = .001 for NMOSD vs. HCs). CONCLUSIONS: The choroid plexus was significantly more enhanced on brain MRI of patients with MS or NMOSD than on that of HCs, suggesting the involvement of the choroid plexus in the autoimmune inflammatory processes in MS and NMOSD.

Serum biomarkers in myelin oligodendrocyte glycoprotein antibody-associated disease.

OBJECTIVE: To test the hypothesis that the pattern of serum biomarkers of disease activity and disability in myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD) will be different from those in neuromyelitis optica spectrum disorder (NMOSD) with anti-aquaporin-4 antibodies (AQP4-Abs). METHODS: Using ultrasensitive single-molecule array assays, we measured neurofilament light chain (NfL), glial fibrillary acidic protein (GFAP), and tau in the sera of consecutive patients with MOGAD (n = 16) and NMOSD with AQP4-Ab (n = 33). Serum biomarker levels were compared between patients in relapse and remission states, and correlations between the levels of these biomarkers and Expanded Disability Status Scale (EDSS) scores were analyzed within each group. RESULTS: In the MOGAD group, the serum tau level was higher in a relapse state than in a remission state (relapse vs remission: 0.5 [0.4-0.5] vs 0.2 [0.1-0.3] pg/mL, p = 0.027). Both serum levels of NfL and tau correlated with the EDSS score (NfL: r = 0.684, p = 0.003; tau: r = 0.524, p = 0.045). Meanwhile, in the NMOSD group, serum NfL and GFAP levels were higher in a relapse state than in a remission state (relapse vs remission: NfL, 34.8 [12.2-62.3] vs 13.0 [11.3-20.0] pg/mL, p = 0.010; GFAP, 253.8 [150.6-303.0] vs 104.4 [93.9-127.9] pg/mL, p = 0.016). Only the serum GFAP level correlated with the EDSS score (r = 0.485, p = 0.012). CONCLUSION: The pattern of serum biomarkers of disease activity and disability in MOGAD showed a distinct feature from those in NMOSD with AQP4-Ab, which implicates different pathogeneses between the 2 diseases.